AI-Enabled Video Diagnosis and Assisted Walking with Exoskeleton for Parkinson’s Disease

Parkinson’s disease is a chronic and progressive disease resulting from the lack of substantia nigra, or a specific type of nerve cell in the brain’s basal ganglia. The four major motor symptoms of Parkinson’s are bradykinesia, rigidity, tremor, and postural or gait instability. This paper seeks to innovate an artificial intelligence (AI) and engineering solution that involves diagnosing different stages of Parkinson’s disease and mediating the external effects of the symptoms on patients’ quality of life. An AI-based approach incorporating video-processing using MediaPipe and pattern recognition with a Long Short-Term Memory (LSTM) neural network diagnoses Parkinson’s patients when stationary. This method mainly focuses on identifying and measuring rest tremors, a typical symptom of Parkinson’s disease. With an analysis of the patient’s disease progression and specific external symptoms, we can assist and train the patient to mitigate the adverse effects of such symptoms.